This revised application is a proposal aimed at determining the potential of preoperative neurobehavioral data as predictors of outcome of anterior temporal lobectomy (ATL), and expanding the traditional end point of seizure outcome to include neurobehavioral consequences (neuropsychological and psychosocial). Predictors will consist of preoperative neuropsychological measures (cognition, memory, language, musical perception, attention; results of intracarotid amytal memory procedures), psychosocial measures (psychopathology, affect, activities of daily living and vocational status), EEG (extra- and intracranial, ictal and interictal) MRI and PET (in a subsample). In a prospective longitudinal design, 30 new patients per year (total n=120) will be evaluated preoperatively and reevaluated 2 weeks, 12 months and then 24 months after surgery. Recently operated patients (n=60) with baseline data will be similarly studied, with a potential follow-up of up to 10 years. They will be compared to non-operated patients with partial epilepsy (n=30) and normal controls (n=30) studied at the same intervals. Systematic examination of interactions between the predictor variables and reorganization underlying outcome. For example, the investigators will test the hypothesis that better postoperative memory and seizure control is obtained in patients with greater specificity of neural lateralization of neurobehavioral and EEG markers and presence of neuropsychological abnormality. The investigators will also attempt to establish objective criteria for neurobehavioral outcome of ATL including the exploration and validity of several new measures.